The present invention generally relates to predicting a vehicular route of travel, and more particularly to predicting a vehicular route of travel without historical route data.
In connected vehicle applications, trajectory prediction of running vehicles may be used in predictive driving assistance services. For example, predictive driving assistance services may provide advanced notification of dynamic road or traffic condition information parking place notification, or commercial location recommendation. As an example, predictive driving assistance services may receive historical data regarding one or more routes that a vehicle has traversed in the past. This data may be referred to as historical route data, and may be used in predicting future vehicle routes of travel. For example, a first route may be traveled by a particular user during weekdays while the user travels to and from a place of employment, while a different second route is commonly traveled by the same user on weekends or holidays when the user is not traveling to work.
As an example, predictive driving assistance services may employ relatively complex and resource intensive analysis of substantially an entire road network when a user travels to an area that they have not traveled to before. Thus, predicting future routes of a vehicle traveling in a new area may be a relatively resource intensive endeavor in the absence of historical route data. Additionally, predicting future routes of a vehicle traveling in a new area may be a relatively slow process in the absence of historical route data, which may result in a delay in alternative route suggestions for a user.